检索结果-内蒙古大学图书馆

deep learning for safety risk management in modular construction: Status, strengths, challenges, and future directions

引用

AUTOMATION IN CONSTRUCTION 2025年 169卷

作者： Junjia, Yin Alias, Aidi Hizami Haron, Nuzul Azam Abu Bakar, Nabilah Univ Putra Malaysia Fac Engn Dept Civil Engn Serdang 43400 Selangor Malaysia

Occupational health risks such as falls from height, electrocution, object strikes, mechanical injuries, and collapses have plagued the construction industry. deep learning algorithms are exploding due to their outstanding analytical capabilities and are believed to improve safety management significantly. Therefore, this paper systematically reviewed the literature on DL algorithms from 2015 to 2024 in modular construction. It found that the six most popular DL algorithms in this area are "Convolutional Neural Network (CNN)," "Recurrent Neural Network (RNN)," "Generative Adversarial Network (GAN)," "Auto-Encoder (AE)," "deep Belief Network (DBN)" and "Transformer." However, in addition to each algorithm's limitations, problems like data constraints, talent gaps, and a lack of guidance frameworks also exist. To address these issues, three strategies are proposed. They are "establishing a multi-modal data sharing platform," "proposing a paradigm framework for the application of DL algorithms," and "constructing a compound construction talent training mechanism," which provide researchers with future references.

关键词： Safety risk Systematic review Modular construction deep learning algorithms Convolutional neural network Autoencoder Generative adversarial network

来源：评论

学校读者我要写书评

暂无评论

An Enhanced Detection System of Autism Spectrum Disorder Using Thermal Imaging and deep learning

引用

SN Computer Science 2025年第2期6卷 1-11页

作者： Jegan Amarnath, J. Meera, S. Department of Computer Science and Engineering Vels Institute of Science Technology & Advanced Studies (VISTAS) Pallavaram Tamil Nadu Chennai 600117 India

Autism Spectrum Disorder (ASD) is a brain-based condition characterized by social difficulties and repetitive behaviors. Traditional diagnostic methods can be subjective and time-consuming. Current diagnostic techniques for ASD often rely on behavioral assessments and clinical observations, which can be inconsistent and prone to subjectivity. Additionally, there is a need for more reliable and objective methods to differentiate between autistic and non-autistic individuals. Existing imaging techniques require further refinement to improve accuracy and efficiency in detecting ASD. The primary objective of this research is to develop and evaluate an enhanced detection system for ASD using thermal imaging and Improved Mask Faster Recurrent Convolutional Neural Network (IMFRCNN). This research focuses on enhancing the detection of ASD using deep learning (DL) methods and thermal imaging. Traditional diagnostic techniques for ASD often rely on subjective behavioral assessments, highlighting the need for more reliable methods. By examining cerebral imaging data from the thermal face image datasets collect facial thermal images using thermal cameras. Ensure the dataset is diverse, including varying expressions and environmental conditions. The study compares the effectiveness of IMFRCNN and the ResNet 50 system in classifying thermal images of 50 autistic and 50 non-autistic individuals. Emotions were evoked using audio-visual stimuli, and temperature variations in facial areas such as the nose, forehead, cheek, and eyes were measured. Anger showed the greatest temperature differences, with a 12.6% variation between autistic and non-autistic individuals. The IMFRCNN achieved a reliability of 96%, while the ResNet 50 achieved 90%. These findings indicate that computer-aided assessment techniques utilizing thermal imaging and DL can provide a consistent and objective method for identifying individuals with ASD. © The Author(s), under exclusive licence to Springer Nature Sing

关键词： deep learning algorithms Diagnostic tool Improved mask faster recurrent convolutional neural network Performance measures

来源：评论

学校读者我要写书评

暂无评论

A Real-Time Multistage False Data Detection Method Based on deep learning and Semisupervised Scoring algorithms

引用

IEEE SYSTEMS JOURNAL 2023年第2期17卷 1753-1764页

作者： Parizad, Ali Hatziadoniu, Constantine J. Southern Illinois Univ Dept Elect & Comp Engn Carbondale IL 62901 USA Virginia Tech Adv Res Inst Blacksburg VA 24061 USA

This article proposes a real-time multistage data-driven method to detect false data injected by an intruder. The proposed method is based on an accurate load model constructed by deep neural networks (long short-term memory and gated recurrent unit) where historical demand along with extra weather features are employed to formulate the problem as a sequence of predictions and build an appropriate DNN model. Adistinct feature of the proposed method that dramatically improves accuracy is hyperparameter optimization (e.g., dropout rate, batch size, beta 1, beta 2, a) based on cross-validation. Since hyperparameter optimization consumes enormous resources to analyze the different combinations of hyperparameters and causes out-of-memory problems, we have implemented our framework based on a parallel distributed computing technique on high-performance computer clusters. A variety of case studies are examined to find the best model. At the end of this step, the best model is employed as a base reference to a semisupervised scoring algorithm for the purpose of finding sequence of injected false data. To improve the detection efficiency, inter-range, pruning, and minimum score are added to the proposed methodology that suppress high false-positive rates. Several scenarios from a real-world dataset have been studied in this article, and the results demonstrate that the proposed method can be used to detect various types of attacks, including scaling, ramping, professional ramping, and random attacks, with a good performance score (i.e., recall, precision, and F1 score) for all of them. Furthermore, it can mitigate false data with the forecasted load values.

关键词： Attack templates deep learning algorithms false data injection attack (FDIA) gated recurrent unit (GRU) hyperparameter optimization semisupervised learning detection method

来源：评论

学校读者我要写书评

暂无评论

deep learning in industrial machinery: A critical review of bearing fault classification methods

引用

APPLIED SOFT COMPUTING 2025年 171卷

作者： Rehman, Attiq Ur Jiao, Weidong Jiang, Yonghua Wei, Jianan Sohaib, Muhammad Sun, Jianfeng E, Shiju Rehman, Khalil Ur Chi, Yongwei Zhejiang Normal Univ Sch Comp Sci & Technol Jinhua 321004 Peoples R China Zhejiang Normal Univ Zhejiang Inst Photoelect Jinhua 321004 Peoples R China Zhejiang Normal Univ Sch Engn Jinhua 321004 Peoples R China Zhejiang Normal Univ Xingzhi Coll Lanxi 321100 Peoples R China Guizhou Univ Key Lab Adv Mfg Technol Minist Educ Guiyang 550025 Guizhou Peoples R China Huazhong Univ Sci & Technol Sch Elect & Elect Engn State Key Lab Adv Electromagnet Engn & Technol Wuhan 430074 Peoples R China Peking Univ Adv Inst Informat Technol AIIT Hangzhou 311215 Peoples R China JIE Holding Grp Co Ltd Hangzhou 311223 Peoples R China

The review provides an overview of the state-of-the-art in deep learning (DL) algorithms for rolling bearing fault classification which remains vital in industrial sectors including transportation, energy, manufacturing, and so forth. Even though they experience a variety of faults, rolling bearings are very crucial in ensuring machine efficiency. This prompts the review of the DL application, which is continuously growing, for intelligently detecting these faults. It comprehensively analyses DL models including Convolutional Neural Networks (CNNs), Auto-Encoders (AEs), deep Belief Neural Networks (DBNs), Recurrent Neural Networks (RNNs), Generative Adversarial Networks (GANs), and some advanced networks i.e., Transfer learning (TL), Transformer Neural Network (TNN), Self-Supervised learning (SSL), Federated learning (FL), Meta-learning and Interpreting Neural Networks assessing their effectiveness and limitations in fault classification. Thus, the current review is unique among available literature at present since it bridges this crucial gap by including all forms of advanced networks and gives an insight into the potential and challenges. Besides, it emphasizes the importance of different sensing techniques and key datasets in the field to show the contribution towards advancements of DL applications. Finally, referring to current challenges and recommendations for future research directions encompassing environmental adaption, sensor deployment, data preprocessing, model training enhancements, algorithm selection, classifier development, and systematic documentation frame the conclusive part of the paper. This

关键词： Rolling Bearing Fault classification methods deep learning algorithms Industrial machinery

来源：评论

学校读者我要写书评

暂无评论

Prediction of cervical spondylotic myelopathy from a plain radiograph using deep learning with convolutional neural networks

引用

European Spine Journal 2025年 1-12页

作者： Tachi, Hiroyuki Kokabu, Terufumi Suzuki, Hisataka Ishikawa, Yoko Yabu, Akito Yanagihashi, Yasushi Hyakumachi, Takahiko Shimizu, Tomohiro Endo, Tsutomu Ohnishi, Takashi Ukeba, Daisuke Sudo, Hideki Yamada, Katsuhisa Iwasaki, Norimasa Hokkaido University Sapporo Japan Department of Orthopaedic Surgery Eniwa Hospital Hokkaido Japan

Purpose: This study aimed to develop deep learning algorithms (DLAs) utilising convolutional neural networks (CNNs) to classify cervical spondylotic myelopathy (CSM) and cervical spondylotic radiculopathy (CSR) from plain cervical spine radiographs. Methods: Data from 300 patients (150 with CSM and 150 with CSR) were used for internal validation (IV) using five-fold cross-validation strategy. Additionally, 100 patients (50 with CSM and 50 with CSR) were included in the external validation (EV). Two DLAs were trained using CNNs on plain radiographs from C3–C6 for the binary classification of CSM and CSR, and for the prediction of the spinal canal area rate using magnetic resonance imaging. Model performance was evaluated on external data using metrics such as area under the curve (AUC), accuracy, and likelihood ratios. Results: For the binary classification, the AUC ranged from 0.84 to 0.96, with accuracy between 78% and 95% during IV. In the EV, the AUC and accuracy were 0.96 and 90%, respectively. For the spinal canal area rate, correlation coefficients during five-fold cross-validation ranged from 0.57 to 0.64, with a mean correlation of 0.61 observed in the EV. Conclusion: DLAs developed with CNNs demonstrated promising accuracy for classifying CSM and CSR from plain radiographs. These algorithms have the potential to assist non-specialists in identifying patients who require further evaluation or referral to spine specialists, thereby reducing delays in the diagnosis and treatment of CSM. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Artificial intelligence Cervical spine radiographs Cervical spondylotic myelopathy Cervical spondylotic radiculopathy Convolutional neural networks deep learning algorithms

来源：评论

学校读者我要写书评

暂无评论

Design and Implementation of Automatic Machine Translation Model Based on deep learning Algorithm

引用

Procedia Computer Science 2025年 261卷 724-733页

作者： Wei Lin School of Economics and Business Foreign Languages Wuhan Technology and Business University Wuhan 430065 Hubei China

As existing machine translation (MT) mainly uses rules or statistical models, it is not effective in dealing with complex languages and diverse contexts, and it is difficult to cope with low-resource language translation and long text context understanding. To solve these problems, this article applies deep learning algorithms into automatic MT models to improve translation quality and efficiency. In the data collection stage, in addition to using the BCC (BLCU Corpus Center) Chinese corpus, the richness of parallel corpora is also increased through crawler tools. In the preprocessing stage, Jieba is used for Chinese word segmentation, and statistics and dictionaries are combined to reduce ambiguity. The Transformer architecture is integrated with the pre-trained BERT (Bidirectional Encoder Representations from Transformers) model to more precisely capture deep semantics, and fine-tune the model to enhance its understanding of specific translation tasks and optimize the interaction between the decoder and the encoder. During training, the Adam algorithm optimization, weight decay, and Dropout strategy are used to prevent overfitting, and mixed precision training is used to improve efficiency. Then, the Horovod framework is used to achieve multi-machine parallel computing. The research results show that the average scores of the MT model using deep learning algorithms on BLEU (Bilingual Evaluation Understudy) and METEOR (Metric for Evaluation of Translation with Explicit ORdering) reach 39.73% and 75.15%, respectively, which is significantly better than the traditional DNN algorithm. At the same time, the model translation takes only 0.36 seconds, which significantly improves efficiency. The model combines pre-training with self-attention mechanism and performs well in the translation of complex sentences and low-resource languages, verifying its accuracy and stability.

关键词： Machine translation models deep learning algorithms transformer architecture pre-trained model translation performance evaluation

来源：评论

学校读者我要写书评

暂无评论

Impact Analysis of Stacked Machine learning algorithms Based Feature Selections for deep learning Algorithm Applied to Regression Analysis

Impact Analysis of Stacked Machine Learning Algorithms Based...

引用

SoutheastCon Conference

作者： Kulkarni, Shrirang Ambaji Gurupur, Varadraj P. King, Christian Univ Cent Florida Sch Global Hlth Management & Informat Orlando FL 32816 USA

ISBN: (纸本)9781665406529

Ensemble learning algorithms have proved to be one of the best machine learning algorithms towards optimal performances in terms of regression and classification tasks for a variety of applications. When applied for small or medium structured datasets, eXtreme Gradient Boosting (XGBoost) has emerged as a popular ensemble learning technique based on its performance merits. In recent years Light Gradient Boosting Machine (LightGBM) has emerged as a promising ensemble strategy that is competing with XGBoost in terms of performance. Also, Lasso Regression has proven capabilities in terms of feature selections and applications for small datasets. This paper illustrates experimentation performed on a diabetes dataset where the authors tested the hypothesis that feature selection has relatively no impact on the performances of deep learning algorithms as they have built-in capabilities in terms of layers to perform feature selection on their own. Therefore, the hypothesis tested stacking using deep learning - Multi-Layer Perceptron (DMLP) with optimal algorithms like XGBoost, LightGBM, and Lasso Regression. In the present work, DMLP with all feature variables (DMLP-ALL) outscored DMLP with stacked selected features (DMLP-MS) by 8.78 % in terms of R-2. Also, DMLP-ALL outperformed the benchmarked algorithm Automated Machine learning (AML) by 10.25% in terms of R-2. The validation of the proposed stacking models by applying a moderate-sized dataset provides promising results for deep learning models stacked with a powerful Level-0 learner.

关键词： Stacking models Machine learning algorithms deep learning algorithms Regression metrics

来源：评论

学校读者我要写书评

暂无评论

TO BE OR NOT TO BE STABLE, THAT IS THE QUESTION: UNDERSTANDING NEURAL NETWORKS FOR INVERSE PROBLEMS

引用

SIAM JOURNAL ON SCIENTIFIC COMPUTING 2025年第1期47卷 C77-C99页

作者： Evangelista, Davide Piccolomini, Elena loli Morotti, Elena Nagy, James g. Univ Bologna Dept Comp Sci & Engn Bologna Italy Univ Bologna Dept Polit & Social Sci Bologna Italy Emory Univ Dept Math Atlanta GA USA

The solution of linear inverse problems arising, for example, in signal and image processing is a challenging problem since the ill-conditioning amplifies, in the solution, the noise present in the data. Recently introduced algorithms based on deep learning overwhelm the more traditional model-based approaches in performance, but they typically suffer from instability with respect to data perturbation. In this paper, we theoretically analyze the trade-off between stability and accuracy of neural networks when used to solve linear imaging inverse problems for cases that are not underdetermined. Moreover, we propose different supervised and unsupervised solutions to increase the network stability and maintain a good accuracy, by means of regularization properties inherited from a model-based iterative scheme during the network training. Extensive numerical experiments on image deblurring confirm the theoretical results and the effectiveness of the proposed deep learning--based approaches to handle noise on the data.

关键词： neural networks stability linear inverse problems deep learning algorithms image deblurring trade-off accuracy stability

来源：评论

学校读者我要写书评

暂无评论

Enhancing surgical precision in squamous cell carcinoma of the head and neck: Hyperspectral imaging and artificial intelligence for improved margin assessment in an ex vivo setting

引用

SPECTROCHIMICA ACTA PART A-MOLECULAR AND BIOMOLECULAR SPECTROSCOPY 2025年 332卷 125817页

作者： Boehm, Felix Alperovich, Anna Schwamborn, Carolin Mostafa, Mayar Giannantonio, Tommaso Lingl, Julia Lehner, Rene Zhang, Xiaohan Hoffmann, Thomas K. Schuler, Patrick J. Ulm Univ Med Ctr Dept Otorhinolaryngol Head & Neck Surg Frauensteige 12 D-89075 Ulm Germany i2SOUL Consortium Surg Oncol Ulm Ulm Germany Carl Zeiss Corp Res & Technol Carl Zeiss Str 22 D-73447 Oberkochen Germany Carl Zeiss Meditec AG Rudolf Eber Str 11 D-73447 Oberkochen Germany Heidelberg Univ Med Ctr Dept Otorhinolaryngol D-69120 Heidelberg Germany

Background: Head and neck cancers, constituting 3-5% of all cancer cases, often require surgical resection for optimal outcomes. Achieving complete resection (R0) is crucial, but current methods, relying on white light endoscopy and microscopy, have limitations. Hyperspectral imaging (HSI) offers potential benefits by capturing detailed spectral information beyond human vision. Material and methods: This study enrolled 32 patients with head and neck squamous cell carcinoma (HNSCC). Following surgical resection specimens underwent ex vivo HSI imaging. Annotated regions were utilized to train a Convolutional Neural Network (CNN) and Graph Neural Network (GNN). Imaging parameters were carefully optimized for efficiency. Results: Our HSI imaging setup required around 12 min per measurement and demonstrated feasibility with promising accuracy. The combination of HSI and artificial intelligence (AI) achieved an 86% accuracy in predicting tumor tissue. Challenges included data volume and extended capture times. Conclusion: Hyperspectral imaging, complemented by AI, shows promise in enhancing tissue differentiation for HNSCC. The study envisions real-time integration of HSI into surgery for margin assessment. Challenges such as data volume and capture times warrant further exploration, emphasizing the need for ongoing investigations to refine clinical applications.

关键词： Head and neck carcinomas Hyperspectral imaging Artificial intelligence Neural network deep learning algorithms Surgical resection Margin assessment Tumor differentiation Squamous cell carcinoma HNSCC

来源：评论

学校读者我要写书评

暂无评论

Control Interface for Multi-User Video Games with Hand or Head Gestures in Directional Key-Based Games

引用

INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS 2025年第1期16卷 49-60页

作者： Ramirez-Valdez, Oscar Baluarte-Araya, Cesar Castillo-Lazo, Rodrigo Ccoscco-Alvis, Italo Valdiviezo-Tovar, Alexander Villafuerte-Quispe, Alexander Zuniga-Huraca, Dylan Univ Nacl San Agustin Arequipa Arequipa Peru

This paper describes the development and implementation of a hand or head gesture-based control interface for video games, enhanced for games that use directional keys. The objective is to develop an adaptive control system for a multiplayer video game that allows users to choose between the use of traditional directional keys or a gesture-based interface. The methodology used follows the Cross-Industry Standard Process for Data Mining (CRISP-DM) development model, which allows a structured integration of analysis, design, implementation and evaluation steps. Technologies such as OpenCV, MediaPipe and deep learning algorithms are used, translating hand movements into directional commands in real time. In addition, the system integrates a client-server architecture based on *** that supports multiple users, enabling an immersive gaming experience on PC and mobile platforms. The results highlight the accuracy of the system and its potential to improve accessibility, especially for users with motor disabilities by using their hands or head movements to control the directional keys. Concluding that the control interface for multi-user video games provides the necessary support to gamers in performing the task, promoting accessibility in the entertainment environment.

关键词： Control interface video games artificial vision gesture-based interface directional commands human-computer interaction deep learning algorithms accessibility real-time pattern recognition

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：